Science AMA Series: I’m Chris San Marchi, a researcher at Sandia National Laboratories. I manage a diverse portfolio of projects related to hydrogen as a zero-emission fuel and diverse energy carrier. Let’s talk hydrogen, the most abundant element in the universe.


Hi Reddit!

I’m Chris San Marchi, a hydrogen and metallurgy researcher on a team at Sandia Labs in California studying how hydrogen and materials interact, and developing codes and standards for hydrogen fueling stations and other energy-related uses of gaseous hydrogen.

Hydrogen, in combination with renewable energy (e.g., wind and solar), is being considered as an alternative to fossil fuels as a zero-emission vehicle fuel as well as to reduce carbon in all energy sectors. Fuel cell electric vehicles are available from major automotive manufacturers and complement battery electric vehicles by providing long range (>300 miles) and rapid refueling (3-5 minutes). Hydrogen fuel cell-powered forklifts have replaced more than 10,000 battery units and growing rapidly, while hydrogen fuel cell auxiliary power units keep phone cell towers operational in the event of natural disaster. The potential for hydrogen to reduce carbon emissions in the energy sector is almost limitless. Hydrogen can be used as a fuel much like natural gas (NG) and is easier to produce than synthetic NG from renewable energy sources (such as solar and wind power) making hydrogen an important storage component to renewable portfolio standards. Hydrogen can also replace carbon-intensive processes in industry, such as for heating and steel manufacturing. It may not be the solution to all of our energy challenges, but hydrogen enables a domestically-produced secure and sustainable energy future.

I will be back at 5 pm ET (2 PT) to answer your questions, ask me anything!

Thank you everyone! We appreciate all of the questions. Special thanks to the Sandia team of experts who contributed today: Joe Ronevich, Mark Allendorf, Becky Levinson, Cliff Hansen, and Tony McDaniel. You can catch up with our work on our website ( or, or follow our updates on twitter (@SandiaLabs).

What is the storage mechanism of choice for hydrogen now? Have pressurized tanks won or are solid storage options like ammonia-borane still being considered? What is the weight penalty for high pressure hydrogen tanks and have the safety people become comfortable with the risks?


Several automakers now have commercially available hydrogen fuel cell cars, all of which store hydrogen as a gas in high-pressure tanks. California is building hydrogen fueling stations and these also store hydrogen as a gas. On this basis, one would say that safety issues have been adequately addressed. It’s too early to say that any storage medium has “won.” However, it is clear that storage in high-pressure tanks cannot meet several DOE targets, including cost and volumetric capacity. Consequently, solid-state options are still being considered at the research level. Ammonia borane has been eliminated from consideration because it requires off-board regeneration. The typical weight penalty assumed by DOE for any storage medium is something like 50%; i.e., in the combined storage system, the “balance of plant” is half of the total weight

I'm sure you've heard Elon Musk's take on hydrogen fuel cell vehicles.

(Edit: That site doesn't allow ad-blockers, so here is an alternative:

I would agree with him. I cannot image hydrogen ever being a practical alternative to batteries for mobile energy storage. And there are additional better alternatives for stationary energy storage, too.

I don't really have a question, but feel free to respond.


“Never” is a long time. Steven Chu, former Secretary of Energy, once remarked that four miracles would be required for hydrogen to become a viable fuel. He has since retreated from that statement. Hydrogen fuel-cell cars are now on the market in the US in limited geographic areas (California in particular), while Japan, Korea, Germany, Denmark and may others are building hydrogen refueling station networks. Hydrogen can be stored at much higher energy densities than batteries, so that makes it more attractive than even state-of-the-art batteries. True, fuel cell cars are just now entering the market, but considering how quickly production of electric vehicles ramped up, it isn’t out of the question that the same could be true of hydrogen-fueled vehicles.

I'm a big fan of hydrogen for future energy needs. How long would it hypothetically take to retrofit the infrastructure necessary for a hydrogen future?


The current infrastructure for natural gas distribution has been in place for many decades and took equally this long to develop. Many of DOE’s target dates are 2050. The reality is that natural gas is still one of the cheapest energy sources and is not going away for the foreseeable future; however, the research for successful development of hydrogen infrastructure is being done now. The infrastructure will be developed partly in response to demand so as hydrogen usage increases, the development of the infrastructure will parallel this.

How do you feel the trajectory of H2 for storage compares to that of batteries? A lot of the argument for why we should use H2 in mobile applications relies on batteries not being good enough. But battery range and recharge time are quickly becoming competitive with what H2 promises for the future (but cannot achieve today). Is the focus for H2 moving to stationary and/or locally mobile applications, like the forklifts you mention?


Hydrogen offers significant advantages technically, but currently is not cost competitive with batteries for many mobile applications. Batteries self-discharge over time and have a limited number of charge/discharge cycles in their lifetime. Hydrogen does not degrade over time and hydrogen storage systems have many more charge and discharge cycles than batteries. Fuel cell power systems weigh significantly less than do batteries. Hydrogen fueling can be done in a matter of a few minutes, e.g., a 5kg H2 tank (roughly 100kWhr) can be filled in 5 minutes at one of the refueling stations in California and many other locations around the world. In contrast, fully charging a 100kWhr battery takes several hours. It is exciting to see hydrogen powered transportation emerging. Besides the Toyota Mirai (on sale near you perhaps?), fleets of fuel-cell powered buses are emerging around the country and world (12 hydrogen fuel cell-powered buses in operation in the Bay Area; Korea is planning 1000s), and there is a hydrogen powered passenger train (the Coradia) planned to begin service in Germany in 2017.

Once you have the hydrogen fuel it is very clean burning but how much energy does it take to produce the hydrogen fuel and how does this compare with the production of other fuel such as gasoline, natural gas, solar, etc?

For example, solar panels must be made, gasoline refined from crude oil that must be pumped out of the ground, natural gas must be captured from underground sources, etc.


This is a very good question. Hydrogen is produced by various methods such as steam methane reformation. The long-term goal is to produce hydrogen form renewable energy sources, such as wind and solar. Hydrogen can be produced directly in some cases (i.e., solar thermochemical processes), or produced by electrolysis of water, ideally using renewable electricity.

What is the state of the art in hydrogen production by alternative means such as photobiological processes (algae, bacteria, ...)?


The DOE Fuel Cell Technologies Office has, and continues to invest in, bio-hydrogen through fermentation and photobiological routes. State of the art R&D for microbes focuses on genetic manipulation to improve H2 production efficiency. The challenges are great because H2 is toxic to most microbes, they tend to make lots of other things besides H2, and they are not equipped with the most efficient light harvesting apparatus. For biomass to H2, current investments are driven towards making methane or more complex fuels from rich and diverse bio-waste streams. Considering “bio-inspired” H2 production research, DOE is heavily invested in artificial photosynthesis. Consult the webpages for the Joint Center for Artificial Photosynthesis (JCAP), and the newly formed EMN consortium on advanced water splitting materials called HydroGen (

Considering the high pressure required for an economical gaseous storage, hydrogen's reactivity and permeability through various materials (such as steel, ebrittlement) would the use and storage of this material requires an increased threat as compared to traditional fuels (propane, gasoline etc)? To compensate wouldn't the general population need to be educated on its use and potential dangers. Is this re-education of the population a significant burden to the implementation of a hydrogen based fuels?


Because the efficiency of hydrogen fuel cell cars is greater than gasoline vehicles, the amount of energy stored on the vehicle is actually less. In the event of an accident, gasoline remains local to the accident and provides a fuel source for a fire. However, hydrogen, because of the buoyancy, disperses quite rapidly making it a lower hazard than gasoline. There are a significant amount of public transportation fuel cell vehicles and buses that are already in operation and therefore the public are already exposed to hydrogen vehicles. Hydrogen fuel cell vehicles are on the road and there are over 20 hydrogen refueling stations in CA ( that are available to the public that provide a very similar experience to the gasoline stations.

Hydrogen leaks through pretty much everything, even solid metal. Can you speak a little around how this problem is handled? Thanks for your AMA.


Hydrogen does permeate metals, but it’s extremely slow. Embrittlement can be an issue, but tanks can be designed to prevent this. New alloys are also being developed that have lower permeability than conventional steels. Concern about leaks from components is sometimes raised, but hydrogen dissipates very rapidly and does not accumulate like other fuels (such as gasoline vapor and LPG).

As I understand it, the hydrogen combines with the oxygen in the air and produces electricity. The waste is mere water that is harmlessly reintroduced into the environment.

As a hypothetical, what would make the following infeasible: "waste" water is retained rather than released into the environment. Solar panels built into the car roof will slowly replenish the fuel by once again splitting the water into hydrogen and oxygen. The system would be simply using the fuel as a battery, except you could still refill with hydrogen should you be too low to wait for recharge.

Thank you for your time!


A closed-loop hydrogen system for a car using solar panels on the vehicle would be environmentally appealing, but with current technologies it is not practical. A fuel cell vehicle travels roughly 70 miles on 1 kg of hydrogen. To produce 1kg of hydrogen per day from a water-splitting electrolyzer using only solar power would require roughly 60m2 of solar panels. That said, there are examples of an all solar-powered car and an all H2 powered house. In fact, such concepts have been around for many years where solar panels on a house provide H2, by electrolyzing water, that is then stored to run both the house and vehicle using fuel cells. The “hydrogen house” project in the US is one such example.

Could hydrogen compete with molten salt for large scale energy storage?


R&D in the concentrated solar power community is focused on how best to store thermal energy from the sun. Salts are designed to store this energy as sensible and latent heat. The energy combined in sensible and latent heat is usually much less than the heat of reaction or combustion of H2. So, the amount of solar energy that can be stored in H2, if it were sourced from splitting water, would vastly exceed the amount that could be stored by salt in an equivalently scaled process. The possibilities exist, but a detailed technoeconomic analysis would be required to decide if one competes well against the other.

What's the advantage of hydrogen storage for electric motors vs just using batteries?


Batteries self-discharge over time and have a limited number of charge/discharge cycles in their lifetime. The cost of battery storage is roughly proportional to the battery capacity. Hydrogen does not degrade over time and in general, the life of hydrogen storage and fuel cells exceeds the life of your vehicle. Hydrogen systems also separate the storage capacity from the fuel cell, which allows scaling of either component to meet the application.

Hi! How much does it currently cost, both monetarily and environmentally, to produce enough hydrogen to propel a mid size vehicle 30 miles?


A fuel cell driven vehicle currently achieves 60-70 miles per kg of H2. The production cost of 1 kg of H2 depends significantly on the feedstock and the process used, but for large-scale steam methane reformation, the production cost is about $2/kg.

How much hydrogen does it take to generate 1kw? How much energy did it take to create the hydrogen?


To generate 1kWhr using a modern fuel cell (60% efficiency from H2 to electricity) requires roughly 5g of H2. The energy required to generate 5g of H2 depends significantly on the feedstock and the process used.

Edit: For production pathways of steam methane reformation of natural gas, the energy in to energy out ratio is ~1.4MJ of NG for 1MJ of Hydrogen. For Coal with Sequestration of the CO2, it’s 1.7MJ Coal to 1MJ of Hydrogen. Any diesel needed for transportation of the hydrogen to its final destination where it will be used will be on top of that.

Then there’s the electricity needed for production. If we’re talking about something centrally produced that needs to be compressed delivered and stored, that’s another .9 MJ in electricity per MJ in Hydrogen. If it’s produced at the use site, there’s more electricity in production, but less in compression, delivery and storage. ~0.3MJ per MJ Hydrogen.

For Electrolysis, production electricity is ~1.5MJ electricity to 1MJ Hydrogen. Compression and storage is another ~0.3MJ per MJ.

1) Hydrogen is very similar to natural gas in terms that they are gases. What are the challenges to storing hydrogen as compared to natural gas considering that the latter is already used in modern verticals?

2) What is/are the blockade to hydrogen not being more commercialised? Is it because big oil lobbies against it?


1) For gaseous storage of hydrogen, some of the challenges are associated with energy density, which requires high pressure or liquefaction. The pressures used in on-board hydrogen storage are 35 or 70 MPa. This is higher than CNG, which is typically less than 25 MPa. The storage tanks however are designed to handle these pressures and therefore risks/challenges are compensated by proper design.

2) H2 is already a commodity chemical in the world today, and over 10 million metric tons are produced in the US annually. The question is not about how to commercialize, it really is about how to popularize and make more use of it as an energy carrier as opposed to commodity chemical. And to do so using renewable energy so that anthropogenic carbon loading (i.e., carbon dioxide) of the environment can be reduced or eliminated.

So how exactly does making Hydrogen into fuel work? Can you explain the process and what's involved?


Molecular hydrogen (H2) by its nature is already a fuel. It can be burned like gasoline or natural gas to provide heat, and thus steam-generated electricity. But in a fuel cell hydrogen is combined with oxygen in air to produce electricity directly. The challenge is that unlike fossil fuels, molecular hydrogen (H2) cannot be mined or extracted directly from the earth. H2 is trapped inside the water molecule (H2O) or the methane molecule (CH4), or even coal. Making H2 means extracting it from these sources, and currently 10 megatons of H2 are made annually in the US, mostly through steam methane reforming, in other words taking H2 from the CH4 molecule.

With any renewable energy option it's worth considering what we still have to dig out of the ground to make it work. With hydrogen itself obviously there's nothing to dig - but what about other things that are needed - catalysts etc, and particularly storage technologies?


Your observation is correct; hydrogen doesn’t occur naturally in its molecular form, so it has to be produced by decomposing something else, like water or methane (i.e., natural gas). This is done on an industrial scale now, but these processes are not carbon neutral and typically produce CO2. Thus, a major challenge is to develop economical carbon-neutral processes to generate H2 – and we can generate hydrogen from renewable electricity. Storage technologies are also an issue, in particular for vehicles. Hydrogen can be stored as a gas in high-pressure tanks, and this is the method used in the new fuel cell vehicles being rolled out by Toyota and Hyundai. However, there are disadvantages to this, including cost (high), design flexibility (low), and volumetric capacity (does not meet DOE targets). Consequently, considerable research is focused on developing solid-state materials that would not have these disadvantages. You mention catalysts; fuel cells use catalysts to react hydrogen with air. Currently, these use costly noble metals such as platinum. Efforts to develop low-cost, efficient catalysts are also underway.

Out of all the zero-emission or carbon reduction projects that involve hydrogen, what would you say has the most potential environmental benefit?

How do you see hydrogen taking part in steel manufacturing? I thought hydrogen causes steel to become brittle if it difuses through the metal or not eliminated from the alloy during the manufacturing process (just going off my welding experience, so sorry if that's an incorrect statement). Thanks!!


Hydrogen can replace coke as an iron ore reduction agent. Coke is very good at reducing iron ore but also has a significant amount of carbon monoxide and carbon dioxide off-product. Steel-making accounts for nearly 10% of all green-house gas (GHG) emission so replacement of coke with hydrogen can result in significant reduction in GHG emissions. You are right that hydrogen can have an embrittling effect in steel but in this does not matter significantly in the molten steel. Often times vacuum or inert gases are filtered through the molten steel during processing to reduce gas species and other impurities. Similarly, the cooling process (if slow enough) will allow sufficient time for the hydrogen to diffuse out of the material… similar to post-weld heat treatment to remove hydrogen. But more research is needed in this area.

Hydrogen may be the most abundant element in the universe, but it is most definitely not the most abundant on earth. How are we beating the challenge of collecting and storing enough hydrogen to power something bigger than a few dozen vehicles?


Correct, H is only the third most abundant element on earth, but unfortunately we cannot make direct use of elemental H. Regarding existing scale of H2 production, 10 million metric tons are produced annually in the US alone. It is essentially a commodity chemical. Toyota, and Hyundai, are producing fuel cell vehicles commercially, with other major car manufacturers following closely behind (Honda and Mercedes). More than 10,000 fuel cell powered fork lifts are in use today in the US, and hundreds of passenger buses have been deployed around the world. So, H2 already powers much more than a few dozen vehicles globally. That said, displacing fossil fuels in the transportation sector is a long term endeavor.

What's the weight/power ratio of a hydrogen cell engine compared to a conventional fuel engine and a battery-powered engine? How will this change in the future?


The comparison with combustion engines is somewhat challenging because a fuel cell generates electricity rather than motive power, so electrical motors and a battery sized for the application are also required. To offer a simplistic answer, one can compare three Toyota vehicles: the Corolla, the Prius Two and the Mirai, with the Mirai. All three offer similar power (130, 121 and 151Hp, respectively); the curb weight for the hydrogen vehicle is currently greater (2800, 3000 and 4000 lbs, respectively). In general however, fuel cell power units are lighter than batteries; for example, fuel cell replacements of battery units in fork lifts are much lighter.

At the current rate of development, how long before we could potentially see an H2 vehicle for the consumer market at non-subsidized, affordable prices (<$20k)?


Argonne National Lab has developed a model called Autonomie where they project vehicle technology evolution out through 2045. Link here: Fuel cell electric vehicle retail prices are projected to reach within ~$2,500 of conventional vehicles (purchase price) by lab year 2045 (market year 2050) in the technology mid-success case.

There has been a lot of talk about helium being lost into space.

Hydrogen is largely protected from that because it is chemically bound to heavier molecules. But if we start using molecular hydrogen on a large scale, could run the risk of losing large amounts of it and permanently altering earths chemical element budget?


The short answer is no. Helium is an inert gas, relatively rare, and is found naturally occurring. Hydrogen although very abundant (3rd most abundant element on earth), is found in compounds and must be separated to use as a fuel. Currently hydrogen gas is produced by steam methane reforming. An alternative source is separating from water through electrolysis, which is very abundant. So in short, there is no risk of running out of hydrogen.

There has been a lot of talk about helium being lost into space.

Hydrogen is largely protected from that because it is chemically bound to heavier molecules. But if we start using molecular hydrogen on a large scale, could run the risk of losing large amounts of it and permanently altering earths chemical element budget?


More importantly, if H2 is stored as fuel and then recombined with O2 to re-create the H2O, then in theory you will have closed the fuel cycle and would greatly reduce risk of H2 leaving the planet.

What are the advantages/disadvantages to producing hydrogen via microbial fermentation? My cursory search didn't turn up too much research in the area and I'm curious why that would be. Thank you for taking the time to do this AMA.


DOE is funding research on microbial generation of hydrogen:

Was there any progress in recent years regarding production of metallic hydrogen? What would be your predictions regarding this?


Hydrogen is a liquid at 20K (very cold) and is one vector for hydrogen distribution. Cryo-compressed hydrogen is another form of hydrogen that is being explored, which is a supercritical fluid. Metallic hydrogen is probably impractical.

What're your thoughts on the Toyota Mirai?

Thanks for doing the AMA. I'm a Sandia Intern based out of Livermore too :)


Thanks for dropping in! As for the Mirai, hope it sells like crazy.

Additional Assets


This article and its reviews are distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and redistribution in any medium, provided that the original author and source are credited.